Movie Recommendation System

Abstract

Recommendation systems have become an essential component in various industries such as e-commerce and OTT-platforms. These systems use various algorithms to recommend the most relevant data to the user. Movie recommendation systems, in particular, suggest movies based on the user's interests, thus saving time and effort for the user in searching through a large list of movies to watch. The aim of this project was to develop a movie recommendation system using cosine similarity algorithm. The system is designed to provide personalized movie recommendations based on the user's movie preferences. The project began with data collection from various sources, including movie reviews, ratings, and user preferences. The collected data was preprocessed and transformed into a structured format suitable for analysis. The development of a cosine similarity algorithm comes next. This algorithm is used to compare two sets of vectors. The technique was used to assess how well the films in the dataset fit the user's preferences. The method for proposing films was built using a web-based interface. The interface allows users to enter their film tastes and receive suggestions based on what they say. The suggestions are presented in descending order of similarity, with the most comparable films at the top. Even films that the user has never heard of could be suggested by the system. Giving users a wide range of recommendations that are tailored to their particular preferences is made possible thanks to this capability. In conclusion, the project's goal of developing a cosine similarity-based movie recommendation system was accomplished. Users could easily access and interact with the system because of its web-based interface, and it was quite accurate at providing individualized recommendations.